[go: up one dir, main page]

WO2020043673A1 - Choix d'un traitement pour un patient - Google Patents

Choix d'un traitement pour un patient Download PDF

Info

Publication number
WO2020043673A1
WO2020043673A1 PCT/EP2019/072737 EP2019072737W WO2020043673A1 WO 2020043673 A1 WO2020043673 A1 WO 2020043673A1 EP 2019072737 W EP2019072737 W EP 2019072737W WO 2020043673 A1 WO2020043673 A1 WO 2020043673A1
Authority
WO
WIPO (PCT)
Prior art keywords
recommendation
follow
patient
processor
finding
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/EP2019/072737
Other languages
English (en)
Inventor
Amir Mohammad TAHMASEBI MARAGHOOSH
Reza SHARIFI SEDEH
Eran RUBENS
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Koninklijke Philips NV
Original Assignee
Koninklijke Philips NV
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Koninklijke Philips NV filed Critical Koninklijke Philips NV
Priority to CN201980063447.4A priority Critical patent/CN112771621A/zh
Priority to US17/272,466 priority patent/US12300391B2/en
Priority to EP19762732.6A priority patent/EP3844764A1/fr
Publication of WO2020043673A1 publication Critical patent/WO2020043673A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H70/00—ICT specially adapted for the handling or processing of medical references
    • G16H70/20—ICT specially adapted for the handling or processing of medical references relating to practices or guidelines
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H10/00—ICT specially adapted for the handling or processing of patient-related medical or healthcare data
    • G16H10/60—ICT specially adapted for the handling or processing of patient-related medical or healthcare data for patient-specific data, e.g. for electronic patient records
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H20/00—ICT specially adapted for therapies or health-improving plans, e.g. for handling prescriptions, for steering therapy or for monitoring patient compliance
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H15/00—ICT specially adapted for medical reports, e.g. generation or transmission thereof
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • G—PHYSICS
    • G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
    • G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
    • G16H40/00—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices
    • G16H40/20—ICT specially adapted for the management or administration of healthcare resources or facilities; ICT specially adapted for the management or operation of medical equipment or devices for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms

Definitions

  • Embodiments described herein generally relate to systems and methods for selecting a treatment for a patient and, more particularly but not exclusively, to systems and methods for selecting a treatment for a patient based on follow-up recommendations to incidental findings.
  • embodiments relate to a method for selecting a treatment for a patient.
  • the method includes extracting, using a processor executing instructions stored on a memory, at least one incidental finding from a record associated with the patient; extracting, using the processor, at least one first follow-up recommendation from the record; associating a first follow-up recommendation with a first incidental finding; obtaining, using an interface, a second follow-up recommendation from a third party-defined policy; determining, using the processor, whether any inconsistencies exist between the first follow-up recommendation and the second follow-up recommendation; presenting, using a user interface, a determination of whether any inconsistencies exist between the first follow-up recommendation and the second follow-up recommendation; receiving, using the user interface, user feedback as to which recommendation is to be used to resolve any determined inconsistencies; and updating, using the processor, a clinical care workflow based at least on the received user feedback.
  • the third party is at least one of a healthcare institution and an administrative body.
  • the method further includes receiving, using the user interface, a resolution upon the processor determining an inconsistency exists between the first follow-up recommendation and the second follow-up recommendation.
  • the at least one incidental finding relates to a finding of a radiology examination.
  • extracting the at least one incidental finding includes: identifying the at least one incidental finding in a medical examination report; and confirming that the at least one incidental finding is not a prior finding by analyzing longitudinal data associated with the patient.
  • the method further includes extracting at least one patient feature from an electronic health record, wherein the at least one patient feature is required by the third party-defined policy, wherein the second follow-up recommendation is based on the at least one extracted patient feature.
  • associating the first follow-up recommendation with the first incidental finding includes using at least one of a keyword matching approach, a model trained on paired training data, and existing guidelines specifying a required incidental finding for a given follow-up recommendation.
  • embodiments relate to a system for selecting a treatment for a patient.
  • the system includes an interface for receiving a record associated with the patient; and a processor executing instructions stored on a memory to extract at least one incidental finding from the record associated with the patient; extract at least one first follow-up recommendation from the record; associate the at least one first follow-up recommendation with the at least one incidental finding; obtain at least one second follow-up recommendation from a third party-defined policy; determine whether any inconsistencies exist between the at least one first follow-up recommendation and the at least one second follow-up recommendation; present, using a user interface, a determination of whether any inconsistencies exist between the at least one first follow- up recommendation and the at least one second follow-up recommendation; receive, using the user interface, user feedback as to which recommendation is to be used to resolve any determined inconsistencies; and update a clinical care workflow based at least on the received user feedback.
  • the third party is at least one of a healthcare institution and an administrative body.
  • the user interface is further configured to receive a resolution upon the processor determining an inconsistency exists between the at least one first follow-up recommendation and the at least one second follow-up recommendation.
  • the at least one incidental finding relates to a finding of a radiology examination.
  • the processor is configured to extract the at least one incidental finding by identifying the at least one incidental finding in a medical examination report; and confirming that the at least one incidental finding is not a prior finding by analyzing longitudinal data associated with the patient.
  • the processor is further configured to extract at least one patient feature from an electronic health record, wherein the at least one patient feature is required by the third party-defined policy, wherein the second follow-up recommendation is based on the at least one extracted patient feature.
  • the processor associates the at least one first follow-up recommendation with the at least one incidental finding using at least one of a keyword matching approach, a model trained on paired training data, and existing guidelines specifying a required incidental finding for a given follow-up recommendation.
  • embodiments relate to a non-transitory computer- readable medium containing computer executable instructions for performing a method for selecting a treatment for a patient.
  • the computer-readable medium includes computer-executable instructions for extracting, using a processor executing instructions stored on a memory, at least one incidental finding from a record associated with the patient; computer-executable instructions for extracting, using the processor, at least one first follow-up recommendation from the record; computer-executable instructions for associating the at least one first follow-up recommendation with the at least one incidental finding; computer-executable instructions for obtaining, using an interface, at least one second follow-up recommendation from a third party-defined policy; computer-executable instructions for determining, using the processor, whether any inconsistencies exist between the at least one first follow-up recommendation and the at least one second follow-up recommendation; computer-executable instructions for presenting, using a user interface, a determination of whether any inconsistencies exist between the at least one first follow up recommendation and
  • FIG. 1 illustrates a system for selecting a treatment for a patient in accordance with one embodiment
  • FIG. 2 depicts a flowchart of a method for selecting a treatment for a patient in accordance with one embodiment
  • FIG. 3 illustrates exemplary follow-up recommendations based on institutional guidelines in accordance with one embodiment.
  • Reference in the specification to“one embodiment” or to“an embodiment” means that a particular feature, structure, or characteristic described in connection with the embodiments is included in at least one example implementation or technique in accordance with the present disclosure.
  • the appearances of the phrase“in one embodiment” in various places in the specification are not necessarily all referring to the same embodiment.
  • the appearances of the phrase“in some embodiments” in various places in the specification are not necessarily all referring to the same embodiments.
  • the present disclosure also relates to an apparatus for performing the operations herein.
  • This apparatus may be specially constructed for the required purposes, or it may comprise a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer or by using some cloud-based solution.
  • a computer program may be stored in a computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, magnetic-optical disks, read-only memories (ROMs), random access memories (RAMs), EPROMs, EEPROMs, magnetic or optical cards, application specific integrated circuits (ASICs), or any type of media suitable for storing electronic instructions, and each may be coupled to a computer system bus.
  • the computers referred to in the specification may include a single processor or may be architectures employing multiple processor designs for increased computing capability.
  • incidental findings do not result in a follow-up recommendation and/or further treatment. This can ultimately lead the patient to go untreated and may further lead to severe health consequences. It is therefore important to provide an automated way to ensure that proper care and attention is provided to incidental findings and the particular ailment(s) associated with the incidental finding.
  • Incidental findings may be common in radiology examinations. There may be three different scenarios related to incidental findings encountered during a radiology examination. In the first scenario, the radiologist may observe the incidental finding and provide a recommendation (e.g., by entering a note of the recommended follow-up in the patient’s health record) that matches guidelines set by the ACR or some other administrative body.
  • a recommendation e.g., by entering a note of the recommended follow-up in the patient’s health record
  • a radiologist may observe the incidental finding and provide a follow-up recommendation.
  • the provided follow-up recommendation does not match the ACR guidelines or guidelines by some institution or administrative body.
  • FIG. 1 illustrates a system 100 for selecting a treatment for a patient in accordance with one embodiment.
  • the system 100 may include a processor 120, memory 130, a user interface 140, a network interface 150, and storage 160 interconnected via one or more system buses 110. It will be understood that FIG. 1 constitutes, in some respects, an abstraction and that the actual organization of the system 100 and the components thereof may differ from what is illustrated.
  • the processor 120 maybe any hardware device capable of executing instructions stored on memory 130 or storage 160 or otherwise capable of processing data.
  • the processor 120 may include a microprocessor, field programmable gate array (FPGA), application-specific integrated circuit (ASIC), or other similar device(s).
  • FPGA field programmable gate array
  • ASIC application-specific integrated circuit
  • the functionality described as being provided in part via software may instead be configured into the design of the ASICs and, as such, the associated software maybe omitted.
  • the processor 120 maybe configured as part of a user device on which the user interface 140 executes or may be located at some remote location.
  • the memory 130 may include various memories such as, for example LI, L2, or L3 cache or system memory. As such, the memory 130 may include static random access memory (SRAM), dynamic RAM (DRAM), flash memory, read only memory (ROM), or other similar memory devices. The exact configuration of the memory 130 may vary as long as instructions for selecting a treatment for a patient can be executed.
  • SRAM static random access memory
  • DRAM dynamic RAM
  • ROM read only memory
  • the user interface 140 may execute on one or more devices for enabling communication with a user such as a clinician or other type of medical personnel.
  • the user interface 140 may include a display, a mouse, and a keyboard for receiving user commands.
  • the user interface 140 may include a command line interface or graphical user interface that may be presented to a remote terminal via the network interface 150.
  • the user interface 140 may execute on a user device such as a PC, laptop, tablet, mobile device, smartwatch, or the like.
  • a user device such as a PC, laptop, tablet, mobile device, smartwatch, or the like.
  • the exact configuration of the user interface 140 and the device on which it executes may vary as along as the features of various embodiments described herein may be accomplished.
  • the user interface 140 may enable a clinician to view imagery related to a medical examination, input notes related to an examination, view notes related to an examination, provide follow-up recommendations to incidental findings, or the like.
  • the network interface 150 may include one or more devices for enabling communication with other hardware devices.
  • the network interface 150 may include a network interface card (NIC) configured to communicate according to the Ethernet protocol.
  • NIC network interface card
  • the network interface 150 may implement a TCP/IP stack for communication according to the TCP/IP protocols.
  • TCP/IP protocols Various alternative or additional hardware or configurations for the network interface 150 will be apparent.
  • the network interface 150 may be in operable communication with one or more sensor devices 151.
  • these may include sensors configured as part of patient monitoring devices that gather various types of information regarding a patient’s health.
  • the one or more sensor devices 151 may include sensors used to conduct a radiology examination.
  • the type of sensor devices 151 used may of course vary and may depend on the patient, context, and the overall purpose of the medical examination. Accordingly, any type of sensor devices 151 may be used as long as they can gather or otherwise obtain the required data as part of an examination.
  • the sensor device(s) 151 may be in communication with the system 100 over one or more networks that may link the various components with various types of network connections.
  • the network(s) may be comprised of, or may interface to, any one or more of the Internet, an intranet, a Personal Area Network (PAN), a Local Area Network (LAN), a Wide Area Network (WAN), a Metropolitan Area Network (MAN), a storage area network (SAN), a frame relay connection, an Advanced Intelligent Network (AIN) connection, a synchronous optical network (SONET) connection, a digital Tl , T3, El , or E3 line, a Digital Data Service (DDS) connection, a Digital Subscriber Line (DSL) connection, an Ethernet connection, an Integrated Services Digital Network (ISDN) line, a dial-up port such as a V.90, a V.34, or a V.34bis analog modem connection, a cable modem, an Asynchronous Transfer Mode (ATM) connection, a Fiber Distributed Data Interface (FDDI) connection, a
  • the network or networks may also comprise, include, or interface to any one or more of a Wireless Application Protocol (WAP) link, a Wi-Fi link, a microwave link, a General Packet Radio Service (GPRS) link, a Global System for Mobile Communication G(SM) link, a Code Division Multiple Access (CDMA) link, or a Time Division Multiple access (TDMA) link such as a cellular phone channel, a Global Positioning System (GPS) link, a cellular digital packet data (CDPD) link, a Research in Motion, Limited (RIM) duplex paging type device, a Bluetooth radio link, or an IEEE 802.11 -based link.
  • WAP Wireless Application Protocol
  • GPRS General Packet Radio Service
  • SM Global System for Mobile Communication G
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple access
  • the storage 160 may include one or more machine-readable storage media such as read-only memory (ROM), random-access memory (RAM), magnetic disk storage media, optical storage media, flash-memory devices, or similar storage media.
  • ROM read-only memory
  • RAM random-access memory
  • magnetic disk storage media magnetic disk storage media
  • optical storage media optical storage media
  • flash-memory devices or similar storage media.
  • the storage 160 may store instructions for execution by the processor 120 or data upon which the processor 120 may operate.
  • the storage 160 may include incidental finding extraction instructions
  • follow-up recommendation extraction instructions 162 follow-up recommendation extraction instructions 162
  • association instructions 163 for associating follow-up recommendations with incidental findings
  • administrative guideline instructions 164 feature extraction and matching instructions 165
  • inconsistency determination instructions 166 user feedback instructions 167
  • clinical care workflow instructions 168 for updating a clinical care workflow.
  • FIG. 2 depicts a flowchart of a method 200 for selecting a treatment for a patient in accordance with one embodiment.
  • Method 200 may rely on the components of the system 100.
  • Step 202 involves extracting, using a processor executing instructions stored on a memory, at least one incidental finding from a record associated with the patient.
  • the processor 120 may perform step 202 by executing the incidental finding extraction instructions 161.
  • the incidental finding extraction instructions 161 may cause the processor 120 to automatically identify and extract all incidental findings given the patient clinical history.
  • These instructions 161 may specifically include identification instructions 169 and classification instructions 170.
  • the identification instructions 169 may be tasked with identifying certain words or other indicia in a patient’s health record and may be executed by an engine consisting of Natural Language Processing (NLP) methods to extract findings based on certain ontologies, word2vec models, regular expressions, etc.
  • NLP Natural Language Processing
  • findings of radiology examinations can be identified and labelled in sentences by keyword-matching using one or more of predefined dictionaries or existing ontologies.
  • These ontologies may include, but are not limited to, SNOMED® and the Unified Medical Language System (UMLS).
  • the incidental findings can be identified and labelled in sentences by using regular expressions to match a pattern or by using previously-labelled data to train any type of appropriate machine learning model.
  • the executed machine learning models may include, but are not limited to, support vector machines, random forests, recurrent neural networks, convolutional neural networks, or any other model that can classify findings from the report.
  • the identified findings are further analyzed to determine whether they are in fact incidental findings and not prior findings.
  • the term“prior findings” may refer to findings that have been observed in prior studies or examinations of the same patient.
  • the processor 120 may execute classification instructions 170 that rely on and use a list of keywords that accompany the incidental findings such as“new” or“incidentally found.”
  • the processor 120 may rely on and use a list of keywords that accompany prior findings such as“previously seen”,“stable”,“unchanged”, or the like. Furthermore, the processor 120 may rely on one or more machine learning procedures such as support vector machines, random forests, recurrent neural networks, convolutional neural networks, or the like, to classify the findings.
  • the identification task and/or the classification task may rely on an ensemble of the approaches described above.
  • the processor 120 may first run keyword matching to identify findings.
  • the processor 120 may train a machine learning model based on examples of such cases in order to identify the remaining findings that are not detected using keyword matching approaches.
  • Step 204 involves extracting, using the processor, at least one first follow-up recommendation from the record.
  • the processor 120 may perform step 204 by executing the follow-up recommendation extraction instructions 162.
  • the follow-up recommendation extraction instructions 162 may enable the processor 120 to automatically detect and extract a follow-up recommendation in a patient health record such as in a clinical report.
  • the follow-up recommendation extraction instructions 162 may include any suitable machine learning procedures and natural language processing techniques that look for specific keywords such as, but not limited to,“follow up” and“recommend.” Additionally or alternatively, the processor 120 may rely on word2vec models trained on a corpus of annotated radiology reports.
  • This follow-up recommendation detection may be implemented as a machine learning problem, provided enough training examples.
  • These machine learning techniques may include, but are not limited to, support vector machines, random forests, or even deep learning approaches such as recurrent neural networks and convolutional neural networks.
  • the processor 120 may execute an ensemble approach that combines keyword-based approaches and machine learning approaches.
  • Step 206 involves associating a first follow-up recommendation with a first incidental finding.
  • the processor 120 may perform step 206 by executing the association instructions 163.
  • the association instructions 163 may link every incidental finding to a potential follow up recommendation.
  • the processor 120 may essentially look for proximity of the occurrence of an incidental finding and the follow-up recommendation phrase(s) and determine whether the follow-up recommendation phrase is related to the incidental finding.
  • the processor 120 may perform this association step in one or more of three ways.
  • the processor 120 may execute keyword-based association instructions 171. According to this technique, the processor 120 may determine the specific finding and the corresponding attributes of the finding in the follow-up recommendation. In radiology-based applications, attributes may include, but are not limited to, nodule size, nodule shape, nodule color, etc. The type of attribute considered may of course vary and may be based on the incidental finding. Then, the processor 120 may match the attributes to the sentence(s) describing the finding.
  • the processor 120 may execute correspondence instructions 172 to prepare paired training data consisting of a finding sentence and the corresponding follow-up sentence. The processor 120 may then provide the paired training data as input to train a machine leaming/deep learning model to determine correspondence.
  • the correspondence instructions 172 may rely on a recurrent neural network model to derive a feature representation for a finding sentence and a feature representation for a follow-up recommendation sentence. Then, the correspondence instructions 172 may rely on a second RNN to derive a co-occurrence feature representation for a finding- follow-up sentence pair.
  • the processor 120 may execute guideline -based instructions 173 for associating incidental findings with follow-up recommendations.
  • the processor 120 may rely on existing guidelines established by administrative bodies such as the Fleischner guidelines (discussed below) to determine which findings require which type of follow up actions, and then look for a corresponding match in a report.
  • Step 208 involves obtaining, using an interface, a second follow-up recommendation from a third party-defined policy.
  • the processor 120 may perform step 208 by executing the administrative guideline instructions 164.
  • the administrative guideline instructions 164 may enable the processor 120 to implement different guidelines for different types of incidental findings.
  • the administrative guideline instructions 164 may include instructions based on ACR guidelines 174 or other standardized guidelines, or instructions based on hospital-specific guidelines 175. These guidelines may be received by the network interface 150 from any appropriate location and/or may be stored in a database accessible by the system 100.
  • the processor 120 may implement any type of rule- based framework such as if-then rules. Additionally or alternatively, the processor 120 may use a machine learning approach to train a model that, by providing a finding and corresponding attributes, can determine which guideline scenario matches best.
  • Step 210 involves extracting at least one patient feature from the electronic health record, wherein the at least one patient feature is required by the third party-defined policy, and the second follow-up recommendation is based on the at least one extracted patient feature.
  • the processor 120 may perform step 210 by executing the feature extraction and matching instructions 165.
  • the processor 120 may execute the feature extraction and matching instructions 165 to extract features that are required based on the administrative guideline instructions 164.
  • the processor 120 may extract these features from, for example, image and/or non-image data from the patient’s record.
  • the features may be extracted from image data (e.g., DICOM® image data) and/or non-image data (EMR data, radiology reports, etc.). Different types of features may of course be extracted from both image and non-image data based on the requirements of the corresponding guidelines.
  • image data e.g., DICOM® image data
  • EMR data e.g., radiology reports
  • Different types of features may of course be extracted from both image and non-image data based on the requirements of the corresponding guidelines.
  • FIG. 3 illustrates a table 300 showing exemplary guidelines based on the ACR guidelines 174.
  • the ACR guidelines are called“Fleischner” guidelines and the required features for extraction are nodule size, smoking history, etc.
  • the processor 120 may extract attributes such as measurement, shapes, margin, or the like.
  • the processor 120 may extract features such as solid/ground, glass/semi-solid, speculated, non-speculated, etc.
  • the instructions 165 for extracting these features may include instructions for extracting these features directly from the image data by using image processing approaches (e.g., segmentation, radiomics, etc.). Additionally or alternatively, attributes may be extracted from a corresponding radiology report by building keyword/regular expression matching methods and/or by training a machine learning model based on training data.
  • image processing approaches e.g., segmentation, radiomics, etc.
  • attributes may be extracted from a corresponding radiology report by building keyword/regular expression matching methods and/or by training a machine learning model based on training data.
  • a clinician or the processor 120 may insert a plurality of ranges (e.g., all possible ranges) of a value for the missing values of the desired features to derive potential ranges of possibilities of outcome. The clinician may then make a decision based on the possible outcomes.
  • the type of nodule is a feature required by the Fleischner Guidelines.
  • the type of nodule may be classified ground-glass, sub-solid, or part-solid. If this information is not available, one can derive the suggested guideline for all three different types of values: Guideline ground glass, Guideline_part_solid, and Guideline sub solid.
  • the feature extraction and matching instructions 165 may then cause the processor 120 to analyze the extracted features to match the features with the administrative guidelines.
  • the extracted features may be fed into an if-then framework, for example, to determine the appropriate follow-up recommendation according to the guidelines.
  • the first step is to determine if the patient is a low risk patient or a high risk patient. This determination is based on the patient’s age, sex, family history, smoking history, etc. This type of data can be automatically extracted from the patient’s electronic medical record data.
  • the ACR guidelines 174 may cause the processor 120 to extract the required data from the patient’s record such as, but not necessarily limited to, nodule size and shape.
  • This data may be extracted directly from an image using image processing techniques (e.g., segmentation) or from a patient’s radiology report using NLP techniques.
  • longitudinal data can be extracted from the history of the patient to determine the required longitudinal information such as nodule growth.
  • the feature extraction and matching instructions 165 may then cause the processor 120 to extract the follow-up sentence(s), word(s), phrase(s) (for simplicity,“sentence(s)”) that are associated with the extracted feature(s).
  • the features extraction and matching instructions 165 may then parse the follow-up sentence(s) to better understand the recommendation. Then, the feature extraction and matching instructions 165 may cause the processor 120 to determine the appropriate follow-up recommendation based on the selected guidelines (e.g., ACR guidelines, hospital guidelines, etc.).
  • Step 212 involves determining, using the processor, whether any inconsistencies exist between the first follow-up recommendation and the second follow-up recommendation.
  • the processor 120 may perform step 212 by executing the inconsistency determination instructions 166.
  • the processor 120 may execute the inconsistency determination instructions 166 to determine whether any inconsistencies exist between the follow-up recommendations provided in the report (i.e., from step 204) and the follow-up recommendations provided by the guidelines (i.e., from step 208). For example, a follow-up recommendation from the report that reads“follow up CT imaging after 6 months” would be inconsistent with a follow-up recommendation from the guidelines that reads“follow up CT imaging after 12 months.”
  • the follow-up recommendation extracted from the report may not include all information that is required to form a“complete” recommendation.
  • a radiologist may provide a follow-up recommendation in the patient’s radiology report such as“a follow-up CT imaging is recommended.” In this scenario, the radiologist forgot to include the timeframe in the recommendation.
  • the inconsistency determination instructions 166 may compare the follow-up recommendation from the guidelines to the “partial” follow-up recommendation to determine the appropriate follow-up recommendation.
  • the appropriate follow-up may be derived based on a maximum match (e.g., modality matching) between the guideline recommendation and the partial recommendation.
  • Step 214 involves presenting, using a user interface, a determination of whether any inconsistencies exist between the first follow-up recommendation and the second follow-up recommendation.
  • the processor 120 may perform step 214 by executing the user feedback instructions 167.
  • the processor 120 may present a message stating that the clinician’s follow-up recommendation and the follow-up recommendation provided by the ACR are in agreement.
  • the processor 120 may present a message warning of any inconsistencies between the follow-up recommendation from the report and the follow-up recommendation provided by, for example, the ACR.
  • the user feedback instructions 167 may convey a message to a clinician to alert them about a partial match.
  • a clinician may review the presented information including the follow-up recommendation and the evidence supporting it, as well as specific image and/or non-image data from which required features were extracted. Furthermore, the clinician can access the reference sentence from where the follow-up recommendation is extracted.
  • Step 216 involves receiving, using the user interface, user feedback as to which recommendation is to be used to resolve any determined inconsistencies.
  • the clinician may select the appropriate follow-up recommendation based on, for example, their knowledge, their past experience, the data supporting the recommended follow-up recommendations, or the like.
  • the clinician may be asked specifically to accept a follow-up recommendation or override a follow-up recommendation.
  • the clinician may view the recommendations (as well as the evidence supporting the recommendations), and provide a response via a user interface such as the user interface 140 of FIG. 1.
  • Step 218 involves updating, using the processor, a clinical care workflow based at least on the received user feedback.
  • the processor 120 may perform step 218 by executing the clinical care workflow instructions 168.
  • the clinical care workflow instructions 168 may enable the processor 120 to prepare the clinical care workflow for the patient based on the selected follow-up. For example, the clinical care workflow instructions 168 may schedule an imaging, schedule a biopsy, report back to the appropriate clinician(s), report back to the patient, or the like.
  • Embodiments of the present disclosure are described above with reference to block diagrams and/or operational illustrations of methods, systems, and computer program products according to embodiments of the present disclosure.
  • the functions/acts noted in the blocks may occur out of the order as shown in any flowchart.
  • two blocks shown in succession may in fact be executed substantially concurrent or the blocks may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
  • not all of the blocks shown in any flowchart need to be performed and/or executed. For example, if a given flowchart has five blocks containing functions/acts, it may be the case that only three of the five blocks are performed and/or executed. In this example, any of the three of the five blocks may be performed and/or executed.
  • a statement that a value exceeds (or is more than) a first threshold value is equivalent to a statement that the value meets or exceeds a second threshold value that is slightly greater than the first threshold value, e.g., the second threshold value being one value higher than the first threshold value in the resolution of a relevant system.
  • a statement that a value is less than (or is within) a first threshold value is equivalent to a statement that the value is less than or equal to a second threshold value that is slightly lower than the first threshold value, e.g., the second threshold value being one value lower than the first threshold value in the resolution of the relevant system.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Epidemiology (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Primary Health Care (AREA)
  • Public Health (AREA)
  • Business, Economics & Management (AREA)
  • General Business, Economics & Management (AREA)
  • Bioethics (AREA)
  • Biomedical Technology (AREA)
  • Measuring And Recording Apparatus For Diagnosis (AREA)

Abstract

L'invention concerne des procédés et des systèmes pour le choix d'un traitement pour un patient. Le système extrait un résultat accessoire d'un dossier associé à un patient et une recommandation de suivi associée. Le système détermine ensuite s'il existe des incohérences entre la recommandation de suivi provenant du dossier et une recommandation de suivi définie par des directives institutionnelles ou administratives. Toutes les incohérences peuvent alors être résolues pour garantir un processus approprié pour la prise en charge du patient.
PCT/EP2019/072737 2018-08-28 2019-08-27 Choix d'un traitement pour un patient Ceased WO2020043673A1 (fr)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201980063447.4A CN112771621A (zh) 2018-08-28 2019-08-27 为患者选择处置
US17/272,466 US12300391B2 (en) 2018-08-28 2019-08-27 Selecting a treatment for a patient
EP19762732.6A EP3844764A1 (fr) 2018-08-28 2019-08-27 Choix d'un traitement pour un patient

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US201862723603P 2018-08-28 2018-08-28
US62/723603 2018-08-28

Publications (1)

Publication Number Publication Date
WO2020043673A1 true WO2020043673A1 (fr) 2020-03-05

Family

ID=67847685

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP2019/072737 Ceased WO2020043673A1 (fr) 2018-08-28 2019-08-27 Choix d'un traitement pour un patient

Country Status (4)

Country Link
US (1) US12300391B2 (fr)
EP (1) EP3844764A1 (fr)
CN (1) CN112771621A (fr)
WO (1) WO2020043673A1 (fr)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20230197276A1 (en) * 2021-03-09 2023-06-22 RAD AI, Inc. Method and system for the computer-assisted implementation of radiology recommendations
US20230343454A1 (en) * 2021-03-09 2023-10-26 RAD AI, Inc. Method and system for the computer-assisted implementation of radiology recommendations
US12354723B2 (en) 2023-04-17 2025-07-08 RAD AI, Inc. System and method for radiology reporting
US12367967B2 (en) 2018-11-19 2025-07-22 RAD AI, Inc. System and method for automated annotation of radiology findings
US12505905B2 (en) 2024-11-19 2025-12-23 RAD AI, Inc. Method and system for the computer-aided processing of medical images

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102261078B1 (ko) * 2019-07-08 2021-06-03 경희대학교 산학협력단 임상 실습 가이드 라인을 컴퓨터 해석 모델로 변환하기 위한 방법 및 이의 시스템
US11342055B2 (en) 2019-09-13 2022-05-24 RAD AI, Inc. Method and system for automatically generating a section in a radiology report
CN115116590B (zh) * 2022-06-29 2023-04-07 中国医学科学院基础医学研究所 深度强化学习方法、装置、肺结节患者随诊流程规划方法、系统、介质和设备

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140365239A1 (en) * 2013-06-05 2014-12-11 Nuance Communications, Inc. Methods and apparatus for facilitating guideline compliance
WO2017064600A1 (fr) * 2015-10-14 2017-04-20 Koninklijke Philips N.V. Systèmes et procédés pour générer des recommandations radiologiques correctes
WO2017077501A1 (fr) * 2015-11-05 2017-05-11 Koninklijke Philips N.V. Profil longitudinal de patient de santé pour des découvertes accidentelles
US20170293734A1 (en) * 2016-04-08 2017-10-12 Optum, Inc. Methods, apparatuses, and systems for gradient detection of significant incidental disease indicators

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8335694B2 (en) * 2004-07-09 2012-12-18 Bruce Reiner Gesture-based communication and reporting system
JP2006259788A (ja) * 2005-03-15 2006-09-28 Seiko Epson Corp 画像出力装置
JP2010525444A (ja) * 2007-04-18 2010-07-22 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ 医療患者に対して私的な逸話を与える装置及び方法
US20080281639A1 (en) 2007-05-09 2008-11-13 Quinn Peter C Real-time evidence- and guideline-based recommendation method and system for patient care
US20130262364A1 (en) 2010-12-10 2013-10-03 Koninklijke Philips Electronics N.V. Clinical Documentation Debugging Decision Support
WO2012104786A2 (fr) * 2011-02-04 2012-08-09 Koninklijke Philips Electronics N.V. Mise à jour de protocole d'imagerie et/ou dispositif de recommandation
US9542532B1 (en) * 2011-10-07 2017-01-10 Cerner Corporation Decision support recommendation optimization
US20130338630A1 (en) * 2012-06-07 2013-12-19 Medtronic Minimed, Inc. Diabetes therapy management system for recommending adjustments to an insulin infusion device
US10902950B2 (en) * 2013-04-09 2021-01-26 Accenture Global Services Limited Collaborative healthcare
US11183300B2 (en) 2013-06-05 2021-11-23 Nuance Communications, Inc. Methods and apparatus for providing guidance to medical professionals
WO2015031296A1 (fr) 2013-08-30 2015-03-05 The General Hospital Corporation Système et procédé destinés à la mise en place d'assistance à la décision clinique pour l'analyse d'imagerie médicale
US20150149215A1 (en) * 2013-11-26 2015-05-28 Koninklijke Philips N.V. System and method to detect and visualize finding-specific suggestions and pertinent patient information in radiology workflow
RU2016140206A (ru) * 2014-03-13 2018-04-13 Конинклейке Филипс Н.В. Система и способ планирования медицинских приемов последующего врачебного наблюдения на основании письменных рекомендаций
CN106415560A (zh) * 2014-06-25 2017-02-15 皇家飞利浦有限公司 辅助患者和临床医师使用共享并且以患者为中心的决策支持工具的系统和方法
CN107408123A (zh) * 2015-02-25 2017-11-28 皇家飞利浦有限公司 用于对临床发现的背景敏感性评价的方法和系统
US20190272919A1 (en) * 2018-03-01 2019-09-05 Siemens Medical Solutions Usa, Inc. Proactive follow-up of clinical findings
US20210142480A1 (en) * 2019-11-12 2021-05-13 Canon Medical Systems Corporation Data processing method and apparatus

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140365239A1 (en) * 2013-06-05 2014-12-11 Nuance Communications, Inc. Methods and apparatus for facilitating guideline compliance
WO2017064600A1 (fr) * 2015-10-14 2017-04-20 Koninklijke Philips N.V. Systèmes et procédés pour générer des recommandations radiologiques correctes
WO2017077501A1 (fr) * 2015-11-05 2017-05-11 Koninklijke Philips N.V. Profil longitudinal de patient de santé pour des découvertes accidentelles
US20170293734A1 (en) * 2016-04-08 2017-10-12 Optum, Inc. Methods, apparatuses, and systems for gradient detection of significant incidental disease indicators

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US12367967B2 (en) 2018-11-19 2025-07-22 RAD AI, Inc. System and method for automated annotation of radiology findings
US20230197276A1 (en) * 2021-03-09 2023-06-22 RAD AI, Inc. Method and system for the computer-assisted implementation of radiology recommendations
US20230343454A1 (en) * 2021-03-09 2023-10-26 RAD AI, Inc. Method and system for the computer-assisted implementation of radiology recommendations
US12354723B2 (en) 2023-04-17 2025-07-08 RAD AI, Inc. System and method for radiology reporting
US12505905B2 (en) 2024-11-19 2025-12-23 RAD AI, Inc. Method and system for the computer-aided processing of medical images

Also Published As

Publication number Publication date
US12300391B2 (en) 2025-05-13
US20210327596A1 (en) 2021-10-21
CN112771621A (zh) 2021-05-07
EP3844764A1 (fr) 2021-07-07

Similar Documents

Publication Publication Date Title
US12300391B2 (en) Selecting a treatment for a patient
CN112365987B (zh) 诊断数据异常检测方法、装置、计算机设备及存储介质
US10818397B2 (en) Clinical content analytics engine
US8612261B1 (en) Automated learning for medical data processing system
US20200185102A1 (en) System and method for providing health information
US8214224B2 (en) Patient data mining for quality adherence
US11651330B2 (en) Machine learning for dynamically updating a user interface
AU2015336146B2 (en) Identification of codable sections in medical documents
US11923094B2 (en) Monitoring predictive models
US20160210426A1 (en) Method of classifying medical documents
US20160110502A1 (en) Human and Machine Assisted Data Curation for Producing High Quality Data Sets from Medical Records
US20230154582A1 (en) Dynamic database updates using probabilistic determinations
Pirneskoski et al. Random forest machine learning method outperforms prehospital National Early Warning Score for predicting one-day mortality: A retrospective study
CN104240171A (zh) 电子病历生成方法及系统
Redfield et al. Derivation and validation of a machine learning record linkage algorithm between emergency medical services and the emergency department
CN117633209A (zh) 用于患者信息概要的方法和系统
Matulionyte et al. Should AI-enabled medical devices be explainable?
US11586955B2 (en) Ontology and rule based adjudication
US20230041051A1 (en) Methods and systems for predicting and preventing frequent patient readmission
CN112329461A (zh) 相似病历确定方法、计算机设备及计算机存储介质
CN111400759A (zh) 访视时间表生成方法及装置、存储介质、电子设备
EP3659150B1 (fr) Dispositif, système et procédé pour optimiser le flux de production d'acquisition d'images
US20220230720A1 (en) Correcting an examination report
CN113688854B (zh) 数据处理方法、装置及计算设备
Woodwark et al. Do blood transfusions make a difference when you are dying?

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19762732

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2019762732

Country of ref document: EP

Effective date: 20210329

WWG Wipo information: grant in national office

Ref document number: 17272466

Country of ref document: US